Research Collection

Conference Paper

SLM Processing of 14 Ni (200 Grade) Maraging Steel

Author(s): Stoll, Philipp; Spierings, Adriaan; Wegener, Konrad; Polster, Stefan; Gebauer, Mathias

Publication Date: 2016

Permanent Link: https://doi.org/10.3929/ethz-a-010802372

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.

ETH Library SLM Processing of 14 Ni (200 Grade) Maraging Steel

Philipp Stoll*1, Adriaan Spierings1, Stefan Polster2, Mathias Gebauer2, Konrad Wegener3

1 inspire AG, St. Gallen, Switzerland 2 Fraunhofer IWU, Chemnitz / Dresden, Germany 3 IWF – Institute for Machine Tools and Manufacturing, ETH Zürich, Zürich, Switzerland

* Corresponding Author: [email protected], +41 71 274 73 27

Abstract Selective Laser Melting (SLM) offers new possibilities in part manufacturing and many of these are of big interest for the tool- and die industry. Besides structural optimisation, further SLM-processable materials are currently in R&D focus, as the combination of structural improvements - such as conformal cooling for tooling applications – the use of application-specific materials can further improve the overall part performance. 14 Ni (200 grade) maraging steel is a die steel that ideally fulfills the requirements of the aluminium die casting process. High thermal conductivity, a low thermal expansion coefficient and stability of mechanical properties at elevated temperatures lead to an amelioration of tool performance. The paper presents a general characterization of 14 Ni (200 grade) maraging steel powder and its SLM processability. A comparison of material properties of additively and conventionally processed 14 Ni (200 grade) maraging steel is shown. The paper finishes with an outlook on further applications for SLM processed 14 Ni (200 grade) maraging steel.

process: high thermal conductivity, low thermal 1 Motivation expansion coefficient and stability of mechanical Selective Laser Melting (SLM) offers new possibilities properties at elevated temperatures. in part manufacturing due to the layerwise build-up These first results are on the general SLM- process. These opportunities are of big interest for many processability, the fabrication of dense parts and the different industrial sectors and their applications, comparison of material properties of conventionally and covering aerospace industry, medical engineering, additively manufactured 14 Ni (200 grade) maraging jewellery and arts as well as the traditional branch of steel. tool- and die-making. Besides structural optimisation and complex design R&D in SLM currently strongly 2 Experimental Setup focuses on the development of further SLM-processable materials [1]. An extension of the list of SLM- 2.1 14 Ni (200 Grade) Maraging Steel Powder processable materials will further push this manufacturing technology. This will also enable the Chemical Composition implementation of this technology in other industrial The 14 Ni (200 grade) maraging steel powder that has sectors, since many industries intend to keep on been processed was produced by gas atomization from working with their approved standard materials and bulk material and delivered by Metso Minerals, Inc., alloys, at least in the beginning. Metso Materials Technology, Finland. The chemical Toolmaking is a very experience based and traditional composition of 14 Ni (200 grade) maraging steel industrial branch but of significant importance for the according to the manufacturer is shown in Table 1. European industrial landscape. While additive tool manufacturing is already state of the art for plastic injection moulding – especially to achieve conformal cooling as shown by Radig [2] – other, metal-working industries currently begin to think about additive manufacturing in general and particularly about SLM. This paper reports on the SLM-processing of 14 Ni (200 grade) maraging steel, commercially known as Marlok® C1650, that – according to its manufacturer [3] – ideally fulfills the requirements of the aluminium die casting Table 1: Chemical composition of 14 Ni (200 grade) . laser power PLaser maraging steel . scan speed vscan . hatch distance h Element wt.% Element wt.% . layer thickness t These four parameters can be put together to the C 0.013 Si 0.057 following equation that calculates the volume energy Cr 0.038 Mn 0.049 density e – i.e. the heat input per volume element – according to [5]. Mo 4.37 S 0.0047 Ni 14.5 P 0.016 P e  Laser (1) Co 10.98 Al 0.056 vscan *h*t Ti 0.24 Cu 0.018 The SLM-experiments were conducted on a Concept Laser M2 machine that is equipped with a fibre laser with maximum output of 194 W at the build platform, Particle Size Distribution operated in cw-mode. A decisive requirement regarding SLM-processability of a powder of some material is its flowability, since it 2.3 Precipitation Heat Treatment enables the powder delivery by the coating device. The The heat treatment has been performed according to the flowability is particularly depending on the particle manufacturer`s guidelines for precipitation heat form and size distribution of the powder. Table 2 treatment [3], which is identical to the heat treatment of displays the data delivered by the powder manufacturer other maraging steels [6, 7]: in volume-% and the data measured at inspire. The latter data covers two values per size range: volume-% and . Heating up to 525°C with maximum heating number-% which enables a more comprehensive rate of 150°C/h understanding of the powder. The data of the . Holding at 525°C for 6 hours manufacturer and inspire match quite well and represent . Cooling down in the furnace according to free a typical distribution of SLM-processable powder. The air cooling analysis of the powder flowability – conducted on Revolution Powder Analyzer, Mercury Scientific Inc., 3 Results and Discussion Newton CT – results in a mean avalanche angle of 38.8 ± 3.5°, and a mean surface fractal of 2.11 ± 0.58. These 3.1 Material Properties values lead – according to the flowability analysis Part Density method developed by Spierings et al. [4] – to a flowability of φ=0.91 which is an excellent value for After the particle size distribution has been evaluated, SLM-processability. other experiments were conducted to check the SLM- processability of 14 Ni (200 grade) maraging steel. The Table 2: Particle size distribution of 14 Ni (200 grade) maraging steel general SLM-processability was analysed by a parameter study to identify the process window for Size range Data Data Data manufacturing of dense and flawless parts, which is a key goal of the SLM-processability and of essential Manufacturer Inspire Inspire importance for applications of SLM-processed 14 Ni Particle dia. Volume - % Volume - % Number - % (200 grade) maraging steel.

< 20 µm 23.23 19.6 49.4 With regard to the part density the volume energy density is a decisive factor. Figure 1 shows the relative 20-32 µm 51.14 50.5 40.7 part density in relation to cast or wrought material as a function of the applied volume energy density during 32-45 µm 25.46 27.4 9.0 the part`s SLM production. The analysis was done > 45 µm 0.17 2.4 1.0 according to Archimedes` principle based on 14 Ni (200 grade) maraging steel raw density of 8.09 g/cm3 that is

given in the manufacturer`s datasheet [3]. The resulting 2.2 SLM Process graph for 14 Ni (200 grade) maraging steel density analysis is shown in Figure 1 and is a characteristic Even though SLM is a highly complex manufacturing trend line for steel powders that are SLM-processed. process and influenced by various physical effects, there Each point in the graph is representing a machine are only four key machine parameters: parameter set. It is clearly visible that a certain amount of energy input is inevitable to achieve acceptable manufacturer. Its density was determined by the relative part densities, i.e. > 99.0%. If higher energy Archimedes` principle which resulted in 8.017 g/cm3. input per volume is applied the relative part density Secondly, this sample was ground to a defined obtained by SLM rises to a range of 99.1 – 99.3%. Since geometry, weighted and then its density was calculated the trend line is not linear there is a limit in achievable to 8.040 g/cm3. Subsequently the values for relative part density. For economic reasons the parameter set for density displayed in Figure 1 – based on raw density of production will be chosen such that sufficient part 8.09 g/cm3 – are minimum values since an evaluation density can be realised with lowest energy input. The based on a raw density below 8.09 g/cm3, as it has been most suitable production parameter – the so-called determined based on bulk material, would result in “standard parameter” – has a volume energy density of higher relative density values. 78.2 J/mm3. The corresponding value is marked with a black arrow in Figure 1, the respective relative density is 99.2%.

Figure 2: Relative part density – 14 Ni (200 grade) Figure 1: Relative part density – 14 Ni (200 grade) maraging steel; density analysis according to maraging steel; density analysis according to Archimedes` principle in comparison to optical analysis Archimedes` principle Despite these uncertainties and deviations regarding the Since the Archimedes` principle is based on a raw raw density of 14 Ni (200 grade) maraging steel it can density value given by the powder manufacturer, further be clearly stated that 14 Ni (200 grade) maraging steel experiments have been conducted to proof this data powder is well processable by SLM, since the provided. An optical assessment of some test samples previously mentioned experiments have successfully has been performed. These samples were cut in their xz- been conducted on another machine setup – Concept planes with z being the direction of SLM build-up. The Laser machine, type M1 – as well. optical evaluation has been conducted on a Keyence Hardness VHX 100 digital microscope with 200x magnification. From each sample, twelve images have been taken The hardness of SLM processed 14 Ni (200 grade) randomly from the respective cross sections, and have maraging steel has been determined by HV10 on the been analysed with regard to pores. In Figure 2 the measuring instrument Brickers 220 from Gnehm arithmetic means of optically analysed parameter sets Härteprüfer AG. The samples were manufactured with are added to the characteristic of relative part density as the standard parameter having a relative density of a function of volume energy density. The optically 99.2%. Hardness analysis is based on eight samples that analysed relative part density is higher for all assessed have been measured five times and in two directions parameter sets, whereat the differences range from respectively, xy-plane that is parallel to the building 0.25% - 0.77%. Spierings et al. [8] showed that with plane and zx-plane beeing vertically oriented to the regard to accuracy the Archimedes` principle has building plane. The value in xy-direction is 360.7 ± 6.0 evident advantages compared to an optical evaluation, HV10, and the value in z-direction is 364.1 ± 6.5 HV10, since the latter is based on the information of one cross respectively. While these results show statistically section whereas Archimedes` principle enables an significant differences based on α=5%, the precipitation overall statement. The comparison of the optically heat treatment that substantially increases the hardness, analysed material density with the Archimedes density leads to an elimination of these differences, since the allows matching the effective bulk material density to a values of 493.0 ± 9.1 HV10 for xy-direction and 491.7 ± 3 value of 8.047 g/cm . Further experiments were 7.2 HV10 for z-direction do not show this statistical conducted to check the raw density of bulk 14 Ni (200 significance anymore. This effect can be explained by grade) maraging steel material, provided by the powder the precipitation that develops isotropic and acts as dislocation barrier in all directions. Static Mechanical Properties For the analysis of the mechanical properties of SLM- processed 14 Ni (200 grade) maraging steel tensile test samples with different specifications have been manufactured. Figure 3 illustrates the variation between the bars: a sample was produced in an upright, vertical orientation while the other two samples were oriented horizontally during the SLM process. For the latter two samples the orientation of the scanning vectors in form of islands relative to the bar axis was divided into two groups. In the first horizontally scanned sample the vectors were parallel and perpendicular to the bar axis – referred to as 0°/90° – whereas the directions changed to +/-45° for the second sample – referred to as +/-45°. Additionally each of these three samples has been Figure 4: Tensile properties of SLM-processed 14 Ni analysed based on five tensile bars respectively in its as- (200 grade) maraging steel in as built and in heat built condition and after the precipitation heat treatment. treated (HT) condition; HT according to the data sheet [3]

Figure 4 shows the results of tensile testing that reveal three notable facts that have been assessed qualitatively so far: 1. The results of the three samples in the as built condition show merely little variance with respect to the orientation of the tensile bars during SLM build-up; the mean values vary by less than 1.3 % for Rp0.2 and 5.3 % for Rm compared to one another. 2. The results of the three samples in the heat treated

condition show merely little variance with respect to the orientation of the tensile bars during SLM build-up; the Figure 3: left: vertically and horizontally oriented mean values vary by less than 3.9 % for R and 3.2 % tensile bars during SLM build-up; right: horizontally p0.2 for R compared to one another. manufactured tensile bar with different orientation of m scanning vectors relative to the bar axis 3. Values of all heat treated samples exceed the mean values of bulk material given in the datasheet [3] by more than 6.7 % for Rp0.2 and 3.4 % for Rm. Observations 1 and 2 are particularly interesting since it is an inherent characteristic of the SLM process to produce parts containing a specific anisotropy due to the columnar microstructure. Regarding the resulting Young’s modulus E that is shown in Figure 5 it is notable that the vertical sample comes up with the lowest value. However, its value in the heat treated condition is in the range of the values given in the datasheet [3]. precipitation heat treatment is subject of ongoing research.

4 Conclusion This paper presents the general SLM-processability of 14 Ni (200 grade) maraging steel powder on two different SLM machine setups. For ongoing investigations only one of these two systems has been selected for the conduction of the experiments. For the chosen machine the ideal process parameter set ensuring part density on one hand and productivity on the other hand requires a volume energy density of 78.2 J/mm3 delivered to the powder bed. Further promising results of SLM-processed 14 Ni (200 grade) maraging steel in comparison to additively manufactured 1.2709 tool steel were achieved and are summarized below: 1. Hardness is in a similar range to 1.2709 tool steel Figure 5: Young’s modulus E of SLM-processed 14 Ni after precipitation heat treatment and it is not depending (200 grade) maraging steel on the build-up direction of the part.

Thermal Conductivity 2. Mechanical properties after heat treatment are superior to the values given in the datasheet [3]. As mentioned in the introduction a beneficial property of 14 Ni (200 grade) maraging steel for its application in 3. The tensile properties show merely little anisotropy in tooling components is its high thermal conductivity. comparison to other SLM-processed steels as shown by Guan et al. [10] and Tolosa et al. [11]. This very interesting result will further be analysed by the authors – particularly based on microstructural images – to get a wider knowledge of the SLM-processability of 14 Ni (200 grade) maraging steel as well as of the SLM process in general since this is a controversial result to current understanding of the process. 4. Thermal conductivity measurements show promising results for an increasing spread of 14 Ni (200 grade) maraging steel as additively manufactured tooling material that unites its beneficial properties with the advantages of the SLM process.

Acknowledgements Figure 6: Thermal conductivity of SLM-processed 14 Ni (200 grade) maraging steel The authors like to thank the project partners in the ManuNet-Project “Hiperformtool”, Ringele AG – Figure 6 shows the results of thermal conductivity Lösungen in Blech, Gut Metallumformung AG and measurements performed by Fraunhofer IWU using the Braun CarTec. In Switzerland the project is co-financed half time method on a Laser Flash Apparatus from by KTI while the German partners are funded by the Netzsch. The chart starts at temperatures of 50 °C since German Federal Ministry of Education and Research the ambiences led to large variations and instabilities in (BMBF) within the Framework Concept ”Research for the measurements in the temperature zone below 50 °C. Tomorrow’s Production”. The arithmetic mean of SLM-processed 1.2709 tool steel [9] in hardened condition is additionally added to Literature the chart for comparison reasons. It is also worth [1] T. Wohlers, “Wohlers Report 2014,” pp. 51-55. mentioning that the measured data for SLM-processed [2] G. Radig, “Werkzeugtechnik sorgt für verzugsfreie 14 Ni (200 grade) maraging steel considerably exceeds komplexe Bauteile,” VDI-Z special Werkzeug- the values provided by the powder manufacturer in the /Formenbau November, 2013, pp. 24-26. datasheet [3] for bulk material. The reasons for the [3] Metso Powdermet Materials Technology substantial increase in thermal conductivity of SLM- Solutions, “Marlok® - Longer die life – Better processed 14 Ni (200 grade) maraging steel due to the quality,” Datasheet, 2005. [4] A.B. Spierings, M. Voegtlin, T. Bauer, K. Wegener, “Powder flowability characterization methodology for powder-bed-based metal additive manufacturing,” Progress in Additive Manufacturing Journal, 2015. [5] B. Vandenbroucke and J.-P. Kruth, “Selective laser melting of biocompatible metals for rapid manufacturing of medical parts,” Rapid Prototyping Journal, vol. 13, no. 4, 2007, pp. 196- 203. [6] M. Cabeza, G. Castro, P. Merino, G. Pena and M. Roman, “A study of laser melt injection of TiN particles to repair maraging tool steels,” Surf. Interface Anal. (2014), ECASIA special issue paper. [7] S. Gulizia, D. Jones, M.Z. Jahedi and P. Koltun, “Thermal Fatigue Behaviour of Die Materials for Diecasting Tooling”, Feature, pp. DC21-DC24. [8] A.B. Spierings, M. Schneider, R. Eggenberger, “Comparison of density measurement techniques for additive manufactured metallic parts,” Rapid Prototyping Journal, Vol. 17, No. 5, 2011, pp. 380- 386. [9] S. Polster, M. Gebauer, Experimental Data, 2015. [10] K. Guan, Z. Wang, M. Gao, X. Li, X. Zeng, “Effects of processing parameters on tensile properties of selective laser melted 304 stainless steel,” Materials and Design, Vol. 50, 2013, pp. 581-586. [11] I. Tolosa, F. Garciandia, F. Zubiri, F. Zapirain, A. Esnaola, “Study of mechanical properties of AISI 316 stainless steel processed by „selective laser melting“, following different manufacturing strategies,” International Journal of Advanced Manufacturing Technology, Vol. 51, 2010, pp. 639- 647.